In May 2023, Surgeon General Vivek Murthy issued an advisory calling loneliness and social isolation a public health epidemic. The report cited research linking social isolation to a 29 percent increased risk of heart disease, a 32 percent increased risk of stroke, and a 50 percent increased risk of developing dementia. These are numbers that belong in a conversation about smoking or obesity — but loneliness rarely gets the same policy attention, because loneliness is invisible and smoking is not.

What the advisory was careful to note — and what subsequent coverage frequently omitted — is that the epidemic is not evenly distributed. It falls harder on men than on women, harder on younger men than older men, harder on men without college degrees than those with them, and hardest of all on men who combine several of these characteristics. The loneliest demographic in America is not, despite the cultural stereotype, the elderly. It is men in their twenties and thirties.

The Data: What Has Actually Changed

The Survey Center on American Life has been tracking male friendship since the 1990s, and its findings are striking. In 1990, 3 percent of American men reported having no close friends. In 2021, that figure was 15 percent — a fivefold increase in three decades. Over the same period, the percentage of men reporting ten or more close friends fell from 40 percent to 28 percent.

This is not a measurement artifact. The trend appears consistently across multiple surveys, multiple methodologies, and multiple countries. A 2023 YouGov survey found that 22 percent of British men said they had no close friends. The Australian Institute of Family Studies has documented similar trends. The pattern is consistent enough to constitute a genuine social change rather than a data quirk.

What changed? Several things simultaneously, which is why the trend has been difficult to reverse. Robert Putnam’s Bowling Alone (2000) documented the first phase of civic disengagement — the collapse of the clubs, unions, churches, and community organizations that had structured male social life through most of the 20th century. These were the places where men had friends by default: you belonged to something, and the belonging meant you had people.

The decline of these institutions accelerated through the 2000s and 2010s. The financial crisis of 2008 hollowed out working-class communities where male friendship had historically been organized around shared industry. The rise of remote work reduced workplace social contact — a primary source of male friendship. Social media provided a simulacrum of connection that satisfied the metric (contacts, follows, likes) while failing to provide the substance (knowing someone, being known, mutual dependence).

Why Men Make Friends Differently — and Why It Matters

The social psychology of male friendship is distinct from female friendship in ways that have practical implications for why male friendships are declining faster. Research consistently shows that men form and maintain friendships primarily through shared activity — doing things together — rather than through the emotional disclosure and conversational intimacy that characterize female friendship.

Stuart Miller’s Men and Friendship (1983) was one of the first serious studies of this phenomenon, and its findings have been replicated repeatedly: men bond at the shoulder, not face to face. The hunting trip, the poker game, the gym session, the round of golf — these are not substitutes for “real” friendship but the actual mechanism by which male friendship forms and is maintained. Take away the shared activity and the friendship frequently dissolves, not because the affection was insincere but because the infrastructure that sustained it is gone.

This makes male friendship structurally fragile in ways that female friendship is not. When a man’s work situation changes, when he moves cities, when he marries and has children — all of the activities that sustained his friendships tend to become unavailable simultaneously. His female partner, who maintains friendship through conversation, can sustain her friendships by phone. He cannot do the equivalent. The friendships atrophy.

The Masculinity Problem

Underlying all of this is a cultural prescription that makes male friendship harder to form and maintain: the prohibition on emotional expression between men. Men who feel affection for other men — who find value in a friendship, who appreciate the other man, who would be devastated to lose the relationship — generally do not say so. The cultural scripting provides no language for this that does not immediately raise questions about sexuality.

This is historically bizarre. Achilles weeps openly for Patroclus in the Iliad. Jonathan and David exchange declarations of love in the Book of Samuel (“your love to me was extraordinary, surpassing the love of women”). Abraham Lincoln wrote letters to his close friend Joshua Speed that would raise eyebrows in contemporary masculinity culture. Victorian men routinely described their male friendships in terms of love, longing, and emotional dependence without any implication of sexuality.

The collapse of this language and permission — largely a product of 20th century homophobia and the intense anxiety it produced around male intimacy — has left men without a way to articulate what their friendships mean to them, which makes those friendships less robust, less able to survive difficulty, and less able to provide the psychological sustenance that close relationships provide.

What the Research Says About Solutions

The good news is that the research on male loneliness is not exclusively grim. Several interventions show consistent effects, and they tend to share common features.

Activity-based communities work. Sports leagues, CrossFit gyms, hiking clubs, music ensembles — environments that give men recurring shared activity with the same group of people — show significant benefits for male social connection. The activity is the point of entry, but the relationship becomes primary over time. A 2022 study in Social Science & Medicine found that men who participated in team sports reported significantly lower loneliness scores than demographically comparable men who did not, even controlling for physical activity levels. The sport was the mechanism, not the exercise.

Deliberate proximity beats curated sociality. The research on how friendships form consistently shows that propinquity — being in physical proximity to people repeatedly, without specific social intention — is the primary driver of friendship formation. The neighbors, the colleagues, the regulars at the same coffee shop. Deliberately engineering this — choosing to live in walkable neighborhoods, working in co-working spaces, attending the same gym classes — is significantly more effective than trying to “make friends” through intentional social events.

Men’s groups work when they provide structure and purpose. Men’s therapeutic groups show good outcomes, but the research also shows that purely therapeutic framing deters many men from joining. Groups that combine purpose (accountability, shared project, mutual improvement) with emotional content show better male recruitment and retention. The ManKind Project, which runs men’s groups with an explicitly masculine framing, has conducted its own outcome research showing improvements in depression, relationship quality, and sense of purpose in participants.

The role of work has changed and men haven’t caught up. For previous generations of working-class men, the workplace was primary male social infrastructure — the factory floor, the trade union meeting, the lunch counter. The decline of manufacturing, the rise of individual contributor work, and the spread of remote work have dismantled this. Men who work remotely are significantly more likely to report loneliness than men who work in person, even when their formal social contact outside work is comparable. This is not obvious — it suggests that the casual, unstructured social contact of workplaces was providing something men did not consciously identify as social connection until it was gone.

What Doesn’t Work

Several responses to the male loneliness crisis are popular and ineffective.

Online community is not a substitute. Discord servers, Reddit communities, gaming clans — these provide social stimulation and can provide genuine meaning, but the research on online-only relationships shows that they do not produce the health and psychological benefits of in-person relationships. Men who report their primary social connection as online report loneliness rates similar to men who report no social connection at all.

Telling men to “open up” without providing infrastructure doesn’t work. The cultural prescription to be more emotionally vulnerable is correct but insufficient without context. Vulnerability requires trust, and trust requires shared history, which requires repeated contact, which requires a mechanism. Telling men to be more open without providing the structural conditions for openness is like telling someone to exercise more without giving them a gym or a trail.

Forcing men into female friendship models doesn’t work. The research consistently shows that attempts to get men to form friendships through emotional disclosure — without the activity layer — fail to recruit or retain most men. This is not because men are emotionally deficient; it is because they are emotionally different, and the difference is real enough to matter for program design. The most successful male friendship interventions work with the male preference for activity-based connection rather than against it.

The Honest Prognosis

The male loneliness crisis will not resolve itself. The structural forces that produced it — the collapse of civic organizations, the decline of manufacturing, the rise of remote work, the evaporation of a shared masculine cultural language for friendship and love — are not going to reverse spontaneously. Fixing it requires deliberate action: by individual men who decide to do the work of maintaining and building friendships, and by the institutions and designers who create the environments in which men meet.

The men who are doing well on this dimension are not doing it accidentally. They have gyms, teams, workshops, bands, poker games, volunteer organizations — structured recurring contexts where they see the same people doing the same thing. They have decided, consciously or not, that friendship is a project that requires tending. This is not complicated, but it is countercultural in a world that has decided that algorithm-mediated connection is sufficient. It isn’t. The data is clear.

Related Reading: